This project utilizes state-of-the-art NMR spectroscopy to study problems that are of continuing interest to the area of environmental health. The primary emphasis involves applications in three areas: 1) understanding how the structural and dynamic behavior of DNA polymerases relates to the fidelity of nucleotide incorporation; 2) studies of ligand-macromolecule interactions; and 3) development and evaluation of new methodologies for the structural and dynamic characterization of proteins and other biological macromolecules in support of the research goals. Progress during the past year on several of these projects is summarized below:[unreadable] [unreadable] Project 1:[unreadable] [unreadable] The E. coli DNA polymerase III system is the main replicative polymerase of this organism and also the most extensively studied model system for elucidating the factors involved in replicative fidelity. We have concluded our study of the structure of the binary complex that involves the catalytic subunit of the proofreading exonuclease epsilon, bound to the phage protein, HOT. HOT is a phage-encoded homolog of the theta-subunit of DNA polymerase III that binds directly to epsilon. However, it became clear that HOT was more amenable to study and we were able to obtain crystals of the HOT-epsilon complex, but not of the theta-epsilon complex. Nevertheless, the sequences of theta and HOT are 50-70% identical, and the structures, which we have previously solved as part of this project, also reveal a very similar, three helix bundle. The &#949;-HOT interface is defined by two areas of contact: an interaction of the previously unstructured N-terminus of HOT with an edge of the &#949; central &#946;-sheet, as well as interactions between HOT and the catalytically important helix &#945;1-loop-helix &#945;2 motif of &#949;. This structure provides insight into how HOT and, by implication, &#952; may stabilize the &#949; subunit thus promoting efficient proofreading during chromosomal replication. Interestingly, we found that the structure of the HOT molecule in the complex with epsilon was more similar to the solution structure of theta than to the structure of HOT itself. This result led to a re-examination of our data for HOT in solution, and to the conclusion that in solution HOT is dimeric. We believe that this dimer structure would be consistent with packaging of the HOT dimer in the phage, and are currently exploring this possibility. [unreadable] [unreadable] Project 2: [unreadable] Nucleotide excision repair (NER) is a major DNA repair mechanism which is highly conserved among all biological systems. Its importance is reflected by the broad substrate range recognized by this system, which includes UV-induced photoproducts, alkylated bases and anti-cancer drug-DNA adducts. In humans, three severe diseases are associated with defects in NER: xeroderma pigmentosum, Cockaynes syndrome and trichothiodystrophy. In prokaryotes, NER is accomplished by three enzymes: Uvr A, B and C. Structural information has been limited by the apparent disorder of the autoregulatory C-terminal domain 4 in crystal structures of intact UvrB; in solution, the isolated domain 4 is found to form a helix-loop-helix dimer. In order to gain insight into the solution behavior of UvrB, we have performed NMR studies on methyl-13Cmethionine-labeled UvrB from B. caldotenax (MW = 75 kD). The 13 methyl resonances were assigned on the basis of site-directed mutagenesis and domain deletion. Methionine 632, located at the potential dimer interface of domain 4, provides an ideal probe for UvrB dimerization behavior. The M632 resonance of UvrB is very broad, consistent with some degree of monomer-dimer exchange, and/or conformational instability of the exposed dimer interface. Upon addition of unlabeled domain 4 peptide, the M632 resonance of UvrB sharpens and shifts to a position consistent with a UvrB-domain 4 heterodimer. A dissociation constant of 3.3 &#956;M for the binding of UvrB with the domain 4 peptide was derived from surface plasmon resonance studies. Due to the flexibility of the domain 3-4 linker, inferred from limited proteolysis data and from the relaxation behavior of linker residue M607, the position of domain 4 is constrained not by the stiffness of the linking segment, but by direct interactions with domains 1-3 in UvrB. In summary, UvrB homodimerization is disfavored, while domain 4 homodimerization and UvrB-domain 4 heterodimerization are allowed.[unreadable] [unreadable] Project 3:[unreadable] Allergens are examples of environmental agents commonly derived from natural products. Allergic reactions occur via various routes of presentation, such as ingestion or inhalation. Among ingested or food allergens, peanut allergies are the most frequent cause of anaphylaxis and anaphylactic death in the US. Ara h 2 is one of the most potent food allergens from peanuts. However, despite its importance in peanut allergey, the only available detailed structural information is a homology model of Ara h 2. We are currently evaluating overexpression protocols in order to produce sufficient quantities for structural characterization. Among inhaled allergens, the most commonly associated with asthma are dust mite allergens. The fact that several major dust mite allergens are proteases has lead to the hypothesis that enzymatic function plays a key role in the sensitization to and/or severity of the allergic response. Der p 5 and Der p 7 are two major allergens from dust mites whose function cannot be derived from sequence information alone. Therefore, structural homologies for these proteins are being sought to provide data on their function. To date, we have overexpressed Der p 5 in abundance and progress is being made on Der p 7. Overexpression is a prerequisite for subsequent structural characterization. This research will hopefully lead to a better understanding of what makes these proteins allergenic, and suggest new treatment possibilities.